Arabidopsis CAM7 and HY5 physically interact and directly bind to the HY5 promoter to regulate its expression and thereby promote photomorphogenesis

Abstract

Arabidopsis thaliana CALMODULIN7 (CAM7), a unique member of the calmodulin gene family, plays a crucial role as a transcriptional regulator in seedling development. The elongated HYPOCOTYL5 (HY5) bZIP protein, an integrator of multiple signaling pathways, also plays an important role in photomorphogenic growth and light-regulated gene expression. CAM7 acts synergistically with HY5 to promote photomorphogenesis at various wavelengths of light. Although the genetic relationships between CAM7 and HY5 in light-mediated seedling development have been demonstrated, the molecular connectivity between CAM7 and HY5 is unknown. Furthermore, whereas HY5-mediated gene regulation has been fairly well investigated, the transcriptional regulation of HY5 is largely unknown. Here, we report that HY5 expression is regulated by HY5 and CAM7 at various wavelengths of light and also at various stages of development. In vitro and in vivo DNA-protein interaction studies suggest that HY5 and CAM7 bind to closely located T/G- and E-box cis-acting elements present in the HY5 promoter, respectively. Furthermore, CAM7 and HY5 physically interact and regulate the expression of HY5 in a concerted manner. Taken together, these results demonstrate that CAM7 and HY5 directly interact with the HY5 promoter to mediate the transcriptional activity of HY5 during Arabidopsis seedling development.

Figure 1.

HY5 Directly Binds to the T/G- and E-Boxes of the HY5 Promoter in Vitro. (A) Diagrammatic representation of a 127-bp (−204 to −330 bp) HY5 promoter fragment containing the T/G- and E-boxes used in the electrophoretic mobility-shift assay. The numbers indicate the distance from the transcriptional start site (TSS). The competition was performed by the 85-bp (−204 to −288 bp) wild type (Wt T/G&E), mutated T/G-box and wild-type E-box (mT/G), mutated E-box and wild-type T/G-box (mE), or mutated T/G- and E-box (mT/G&E), as shown. The wild-type and mutated versions (in red) of the T/G- and E-box sequences are given in the boxes. (B) Gel-shift assays using GST-HY5 and the 127-bp T/G- and E-boxes containing the HY5 promoter as probe. No protein was added in lane 1, and ∼500 ng of GST protein was added in lane 2. In lanes 3 to 11, ∼300 ng of GST-HY5 protein was added. Competition was performed with 50 (lanes 4, 6, 8, and 10) and 100 (lanes 5, 7, 9, and 11) molar excess of wild-type or mutated versions of the 85-bp DNA fragment of the HY5 promoter, as shown by the triangles. The plus and minus signs indicate presence and absence, respectively. (C) Gel-shift assays using GST-HY5 and 85-bp wild-type or various mutated versions of the T/G- and E-boxes (as shown in [A]) containing HY5 promoter as probe. No protein was added in lane 1, and ∼500 ng of GST protein was added in lane 2. In lanes 3, 5, 7, and 9, ∼300 ng of GST-HY5 was added, and in lanes 4, 6, 8 and 10, ∼600 ng was added. (D) ChIP assays of the HY5 promoter from Col-0, HY5OE, and hy5 mutant seedlings using antibodies to HY5. The light-inducible NIA2 promoter fragment that does not contain any T/G- or E-box was used as a control (Kushwaha et al., 2008). The results of quantitative real-time PCR are presented as the ratio of the amount of DNA immunoprecipitated from Col-0, HY5OE, or hy5 mutants to input DNA from various backgrounds. The error bars indicate sd of three technical replicates. The experiment was repeated three times, and a representative result is shown.

Figure 2.

Effect of the hy5 Mutation on the Wavelength-Specific Activation of the HY5 Promoter. In each panel of (A), wild-type seedlings are shown on the left and hy5 mutant seedlings are shown on the right. The quantitative GUS activities in (B) to (G) are averages of four technical repeats in one representative experiment (out of three), and the error bars indicate sd. (A) Five-day-old seedlings carrying the ProHY5-GUS transgene were grown in constant darkness (dark), WL (80 μmol m⁻² s⁻¹), BL (30 μmol m⁻² s⁻¹), RL (40 μmol m⁻² s⁻¹), or FR (30 μmol m⁻² s⁻¹). The promoter activity in wild-type and hy5-215 mutant backgrounds was estimated by quantitative GUS activity staining for the same length of time. Bar = 1 mm. (B) and (C) GUS activity of 5-d-old constant dark–grown (B) or constant WL–grown (C) seedlings. (D) GUS activity of 4-d-old dark-grown seedlings transferred to WL for 12, 24, and 48 h. (E) to (G) GUS activity of 5-d-old BL-grown (E), RL-grown (F), and FR-grown (G) seedlings. The wild type and cam7-1 and hy5-215 mutants used are in the Col-0 background.

Figure 3.

Effect of the hy5 Mutation on the Tissue-Specific Expression of the ProHY5-GUS Transgene. In each panel, wild-type seedlings are shown on the left and hy5-215 mutant seedlings are shown on the right. (A) and (B) Leaves (A) and roots (B) of 15-d-old dark-grown plants carrying the ProHY5-GUS transgene. (C) to (E) Leaves (C), stems (D), and roots (E) of 15-d-old WL-grown (16-h-light/8-h-dark cycle) plants carrying the ProHY5-GUS transgene. (F) to (I) Leaves (F), stems (G), roots (H), and flowers (I) of 30-d-old WL-grown (16-h-light/8-h-dark cycle) plants carrying the ProHY5-GUS transgene. (J) and (K) Quantification of GUS activities in leaves, stems, roots, and flowers. Promoter activities were monitored by measuring the GUS activities of wild-type and hy5 mutant seedlings carrying the ProHY5-GUS transgene. The quantitative GUS activities are averages of four independent repeats in one representative experiment (out of four), and the error bars indicate sd. Comparison of GUS activities in roots, stems, and leaves of 15-d-old light-grown (16-h-light/8-h-dark cycle) plants of the wild type versus hy5 mutants is shown in (J), and comparison of GUS activities in roots, stems, leaves, and flowers of 30-d-old light-grown (16-h-light/8-h-dark cycle) plants of the wild type versus hy5 mutants is shown in (K).

Figure 4.

CAM7 Directly Binds to the T/G- and E-Boxes of the HY5 Promoter in Vitro. (A) Gel-shift assays using GST-HY5 and the 127-bp T/G- and E-boxes containing the HY5 promoter as probe. No protein was added in lane 1, and ∼500 ng of GST was added in lane 2. In lanes 3 to 11, ∼300 ng of GST-CAM7 was added. The competition was performed using the 85-bp (−204 to −288 bp) wild type (Wt T/G&E), mutated T/G-box and wild-type E-box (mT/G), mutated E-box and wild-type T/G-box (mE), or mutated T/G- and E-boxes (mT/G&E), as shown in Figure 1A. Competition was performed with 50 (lanes 4, 6, 8, and 10) and 100 (lanes 5, 7, 9, and 11) molar excess of wild-type or mutated versions of an 85-bp DNA fragment of the HY5 promoter, as shown by the triangles. The plus and minus signs indicate presence and absence, respectively. The arrowhead indicates the DNA-protein complex. (B) Gel-shift assays using GST-CAM7 and the 85-bp wild-type or various mutated versions of T/G- and E-boxes containing the HY5 promoter as probe. No protein was added in lane 1, and ∼500 ng of GST was added in lane 2. In lanes 3, 5, 7, and 9, ∼300 ng of GST-CAM7 protein was added, and in lanes 4, 6, 8, and 10, ∼600 ng was added. The arrowhead indicates the DNA-protein complex. (C) Gel-shift assays using GST-CAM7 or GST-CAM7-M2 protein and the 127-bp wild-type HY5 promoter as probe. No protein was added in lane 1, and ∼500 ng of GST was added in lane 2. In lanes 3 and 4, ∼250 and 500 ng of GST-CAM7 was added, respectively; in lanes 5, 6, and 7, ∼250 ng, 500 ng, and 1 μg of GST-CAM7-M2 was added, respectively. The arrowhead indicates the DNA-protein complex. (D) ChIP assays of the HY5 promoter from wild-type (Col-0), CAM7OE, and CAM3OE transgenic seedlings using antibodies to cMyc. The light-inducible NIA2 promoter fragment that does not contain any T/G- or E-box was used as a control (Kushwaha et al., 2008). The results of quantitative real-time PCR are presented as the ratio of the amount of DNA immunoprecipitated from Col-0, CAM7OE, or CAM3OE to input DNA from various backgrounds. The error bars indicate sd of three technical replicates. The experiment was repeated three times, and a representative result is shown.

Figure 5.

Effect of CAM7 on Wavelength-Specific Activation of the HY5 Promoter. In (A) to (E), wild-type, CAM7OE, and cam7 seedlings carrying the ProHY5-GUS transgene are shown from left to right, respectively. The quantitative GUS activities in (F) are averages of four independent repeats in one representative experiment (out of three), and the error bars indicate sd. (A) to (E) Five-day-old seedlings carrying the ProHY5-GUS transgene were grown in constant darkness (A), WL (30 μmol m⁻² s⁻¹ [B]), RL (30 μmol m⁻² s⁻¹ [C]), BL (30 μmol m⁻² s⁻¹ [D]), or FR (30 μmol m⁻² s⁻¹ [E]) for GUS activity staining. Bar = 1 mm. (F) GUS activities of 5-d-old seedlings grown in constant darkness or various light conditions.

Figure 6.

Effect of the cam7 Mutation on the Tissue-Specific Expression of the ProHY5-GUS Transgene. In (A) to (G), wild-type, CAM7OE, and cam7 seedlings carrying the ProHY5-GUS transgene are shown from left to right, respectively. The quantitative GUS activities in (H) are averages of four independent repeats in one representative experiment (out of three), and the error bars indicate sd. (A) to (C) Roots (A), stems (B), and leaves (C) of 15-d-old WL-grown (16-h-light/8-h-dark cycle) plants carrying the ProHY5-GUS transgene. (D) to (G) Roots (D), stems (E), leaves (F), and flowers (G) of 30-d-old WL-grown (16-h-light/8-h-dark cycle) plants carrying the ProHY5-GUS transgene. (H) Comparison of GUS activities in roots, stems, leaves, and flowers of 30-d-old WL-grown (16-h-light/8-h-dark cycle) plants.

Figure 7.

CAM7 Physically Interacts with HY5. (A) In vitro binding of HY5 and CAM7. Approximately 2 µg of CAM7-6His or COP1-6His was individually bound to Ni-NTA beads. GST-HY5 or GST protein was added in equimolar ratio. Supernatant and pellet fractions were fractionated by SDS-PAGE, blotted, and probed with anti-GST antibodies. Lanes 1 and 6 show COP1-6His with GST-HY5 (positive control), lanes 2 and 5 show CAM7-6His with GST-HY5, and lanes 3 and 4 show CAM7-6His with GST. (B) In vitro pull-down assays of cMyc-CAM7 and GST-HY5. GST and GST-HY5 proteins (1 µg each) were individually bound to GST beads. Approximately 1 mg of total protein extract from the CAM7-cMyc overexpression line was added and incubated at 4°C for 4 h. The supernatant and pellet were fractionated by SDS-PAGE, blotted, and probed with anti-cMyc antibodies. (C) and (D) CAM7 colocalizes and interacts with HY5 in the nucleus of onion epidermal cells. In both panels, image (a) shows CFP channel fluorescence, changed to red color, (b) shows YFP channel fluorescence, changed to green color, and (c) shows merged images of (a) and (b). (E) and (F) Empty BiFC vectors (E) and CAM7-YFPN-ter and HY5-YFPC-ter constructs (F) were cotransformed into onion epidermal cells. In both panels, image (a) shows 4′,6-diamidino-2-phenylindole (DAPI) fluorescence for the confirmation of nuclei, (b) shows the YFP channel image produced by reconstruction of YFP, (c) shows merged images of (a) and (b), and (d) shows the respective bright-field image (differential interference contrast [DIC]). Arrows indicate the positions of the nuclei.

Figure 8.

CAM7 and HY5 Together Bind to the HY5 Promoter to Regulate HY5 Expression. (A) In vivo interactions between CAM7 and HY5 proteins in ChIP assays. The cross-linked complex of CAM7/HY5 with the HY5 promoter was pulled down by antibodies to cMyc. The complex was reverse cross-linked and resolved by SDS-PAGE. Both the input and immunoprecipitates were probed with antibodies to HY5. (B) DNase I footprinting analysis of the 127-bp HY5 promoter fragment. Lane 1 shows the DNase I cleavage pattern with 20 μg of GST protein. Lanes 2 to 4 show the DNase I cleavage pattern with 20, 20, and 20 + 20 μg of the GST-HY5, GST-CAM7, and GST-HY5 + GST-CAM7 proteins, respectively. The protected DNA sequence containing the T/G- and E-box of the HY5 promoter fragment is shown at right. (C) ChIP assays of the HY5 promoter from Col-0, CAM7OE, and hy5 CAM7OE seedlings using antibodies to HY5. For experimental details, see the legend to Figure 1D. *P < 0.001 (Student’s t test). (D) Real-time PCR analysis of HY5 transcript levels from 4-d-old WL-grown seedlings (30 μmol m⁻² s⁻¹). ACTIN2 was used as a control. Error bars represent sd. n ≥ 3 independent experiments with similar results. *P < 0.01 (Student’s t test). (E) Working model of CAM7- and HY5-mediated regulation of HY5. HY5 and CAM7 bind to the HY5 promoter to promote HY5 expression. The accumulated HY5 protein then binds to a large number of promoters to activate multiple signaling and metabolic pathway genes.